@inproceedings{3a40c1d927a34cb6869acf40a0accc32,
title = "Jigsaw: A High-Utilization, Interference-Free Job Scheduler for Fat-Tree Clusters",
abstract = "Jobs on HPC clusters can suffer significant performance degradation due to inter-job network interference. Approaches to mitigating this interference primarily focus on reactive routing schemes. A better approach - -in that it completely eliminates inter-job interference - -is to implement scheduling policies that proactively enforce network isolation for every job. However, existing schedulers that allocate isolated partitions lead to lowered system utilization, which creates a barrier to adoption. Accordingly, we design and implement Jigsaw, a new job-isolating scheduling approach for three-level fat-trees that overcomes this barrier. Jigsaw typically achieves system utilization of 95-96%, while guaranteeing dedicated network links to jobs. In scenarios where jobs experience even modest performance improvements from interference-freedom, Jigsaw typically leads to lower job turnaround times and higher throughput than traditional job scheduling. To the best of our knowledge, Jigsaw is the first scheduler to eliminate inter-job network interference while maintaining high system utilization, leading to improved job and system performance.",
keywords = "fat-tree, inter-job network interference, scheduling, utilization",
author = "Smith, {Staci A.} and Lowenthal, {David K.}",
note = "Publisher Copyright: {\textcopyright} 2020 ACM.; 30th International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2021 ; Conference date: 21-06-2021 Through 25-06-2021",
year = "2021",
month = jun,
day = "21",
doi = "10.1145/3431379.3460635",
language = "English (US)",
series = "HPDC 2021 - Proceedings of the 30th International Symposium on High-Performance Parallel and Distributed Computing",
publisher = "Association for Computing Machinery, Inc",
pages = "201--213",
booktitle = "HPDC 2021 - Proceedings of the 30th International Symposium on High-Performance Parallel and Distributed Computing",
}